Generative Artificial Intelligence in Cybersecurity: Applications, Opportunities, Risks, Governance, and Future Directions
DOI:
https://doi.org/10.65336/WJMS.2026.31004Keywords:
Generative artificial intelligence, Large language models, Cybersecurity, Cyber threat intelligence, Prompt injection, Vulnerability assessment, Security operations, AI governance, Agentic AI, Human–AI collaborationAbstract
Generative Artificial Intelligence (GenAI), and especially Large Language Models (LLMs), is changing how cybersecurity work is done, because these systems can produce and reshape natural language, code, and other digital material. In this review, we bring together recent academic and institutional sources, concentrating on the period 2024–2026, to analyze how GenAI is applied in cybersecurity, what opportunities and risks it brings, how it is governed, and where research is still lacking. Reported uses include cyber threat intelligence, threat detection, vulnerability assessment and remediation, defense against phishing and social engineering, penetration testing, deception, security operations, incident response, and software security. GenAI is, however, a dual-use technology. Empirical work indicates that it can make phishing more effective and cheaper to run, help attackers exploit known vulnerabilities, and open new attack surfaces such as indirect prompt injection and the poisoning of training data. Our analysis suggests that GenAI is most valuable when it supports expert judgment instead of substituting for it. Its adoption is held back by prompt injection, hallucination, leakage of sensitive information, data and model poisoning, weak supply chains, excessive agency, and evaluation practices that differ from study to study. When we compare the main governance instruments—NIST guidance, the OWASP Top 10 for LLM applications, ENISA and CISA publications, the EU AI Act, and ISO/IEC 42001—we find that they agree in principle but are only loosely connected in practice. Drawing on these findings, we propose a four-stage Human–AI Collaboration Model (GenAI assistance, human validation, controlled action, and continuous feedback) that turns the evidence into a workable process for security teams. We close with priorities for standardized benchmarking, secure agentic systems, privacy-preserving deployment, and validation in real environments.
References
1. Alam, M. T., Bhusal, D., Nguyen, L., & Rastogi, N. (2024). CTIBench: A benchmark for evaluating LLMs in cyber threat intelligence. In Advances in Neural Information Processing Systems (Vol. 37, pp. 50805–50825). Curran Associates.
2. Alqahtani, H., & Kumar, G. (2026). Large language models for cybersecurity intelligence: A systematic review of emerging threats, defensive capabilities, and security evaluation frameworks. Computers, Materials & Continua, 87(3). https://doi.org/10.32604/cmc.2026.077367
3. Balasubramanian, P., Liyana, S., Sankaran, H., Sivaramakrishnan, S., Pusuluri, S., Pirttikangas, S., et al. (2025). Generative AI for cyber threat intelligence: Applications, challenges, and analysis of real-world case studies. Artificial Intelligence Review, 58, Article 336. https://doi.org/10.1007/s10462-025-11338-z
4. Bhatt, M., Chennabasappa, S., Nikolaidis, C., Wan, S., Evtimov, I., et al. (2023). Purple Llama CyberSecEval: A secure coding benchmark for language models (arXiv:2312.04724). arXiv. https://doi.org/10.48550/arXiv.2312.04724
5. Carlini, N., Jagielski, M., Choquette-Choo, C. A., Paleka, D., Pearce, W., Anderson, H., Terzis, A., Thomas, K., & Tramèr, F. (2024). Poisoning web-scale training datasets is practical. In 2024 IEEE Symposium on Security and Privacy (SP) (pp. 407–425). IEEE.
6. Chen, Y., Cui, M., Wang, D., Cao, Y., et al. (2024). A survey of large language models for cyber threat detection. Computers & Security, 145, Article 104016. https://doi.org/10.1016/j.cose.2024.104016
7. Chowa, S. S., Alvi, R., Rahman, S. S., Rahman, M. A., Raiaan, M. A. K., Islam, M. R., Hussain, M., & Azam, S. (2026). From language to action: A review of large language models as autonomous agents and tool users. Artificial Intelligence Review, 59, Article 71. https://doi.org/10.1007/s10462-025-11471-9
8. Cybersecurity and Infrastructure Security Agency [CISA]. (2023). Roadmap for artificial intelligence. U.S. Department of Homeland Security.
9. Das, B. C., Amini, M. H., & Wu, Y. (2025). Security and privacy challenges of large language models: A survey. ACM Computing Surveys, 57(6), Article 152. https://doi.org/10.1145/3712001
10. Deng, G., Liu, Y., Mayoral-Vilches, V., Liu, P., Li, Y., Xu, Y., Zhang, T., Liu, Y., Pinzger, M., & Rass, S. (2024). PentestGPT: Evaluating and harnessing large language models for automated penetration testing. In Proceedings of the 33rd USENIX Security Symposium (pp. 847–864). USENIX Association.
11. European Parliament and Council of the European Union. (2024). Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act). Official Journal of the European Union, L 2024/1689.
12. European Union Agency for Cybersecurity [ENISA]. (2024). ENISA threat landscape 2024. ENISA.
13. Fang, R., Bindu, R., Gupta, A., & Kang, D. (2024). LLM agents can autonomously exploit one-day vulnerabilities (arXiv:2404.08144). arXiv. https://doi.org/10.48550/arXiv.2404.08144
14. Ferrag, M. A., Alwahedi, F., Battah, A., Cherif, B., Mechri, A., Tihanyi, N., Bisztray, T., & Debbah, M. (2025). Generative AI in cybersecurity: A comprehensive review of LLM applications and vulnerabilities. Internet of Things and Cyber-Physical Systems, 5, 1–46. https://doi.org/10.1016/j.iotcps.2025.01.001
15. Greshake, K., Abdelnabi, S., Mishra, S., Endres, C., Holz, T., & Fritz, M. (2023). Not what you've signed up for: Compromising real-world LLM-integrated applications with indirect prompt injection. In Proceedings of the 16th ACM Workshop on Artificial Intelligence and Security (pp. 79–90). ACM. https://doi.org/10.1145/3605764.3623985
16. Heiding, F., Schneier, B., Vishwanath, A., Bernstein, J., & Park, P. S. (2024). Devising and detecting phishing emails using large language models. IEEE Access, 12, 42131–42146. https://doi.org/10.1109/ACCESS.2024.3375882
17. Hu, Y., Zou, F., Han, J., Sun, X., & Wang, Y. (2024). LLM-TIKG: Threat intelligence knowledge graph construction utilizing large language model. Computers & Security, 145, Article 103999. https://doi.org/10.1016/j.cose.2024.103999
18. Ibrar, W., Mahmood, D., Al-Shamayleh, A. S., Ahmed, G., Alharthi, S. Z., & Akhunzada, A. (2025). Generative AI: A double-edged sword in the cyber threat landscape. Artificial Intelligence Review, 58, Article 285. https://doi.org/10.1007/s10462-025-11285-9
19. International Organization for Standardization & International Electrotechnical Commission [ISO/IEC]. (2023). Information technology — Artificial intelligence — Management system (ISO/IEC Standard No. 42001:2023). ISO.
20. Iturbe, E., Llorente-Vázquez, O., Rego, A., Rios, E., & Toledo, N. (2024). Unleashing offensive artificial intelligence: Automated attack technique code generation. Computers & Security, 147, Article 104077. https://doi.org/10.1016/j.cose.2024.104077
21. Jangirala, S., Gupta, V., & Mandal, A. (2026). Large language models in cybersecurity: A PRISMA-ScR-guided scoping review of threats, defenses, and emerging applications. IEEE Access, 14, 117961–117990. https://doi.org/10.1109/ACCESS.2026.3719163
22. Johnson, S., Hassing, R., Pijpker, J., & Loves, R. (2024). A modular generative honeypot shell. In 2024 IEEE International Conference on Cyber Security and Resilience (CSR). IEEE. https://doi.org/10.1109/CSR61664.2024.10679411
23. Karras, A., Theodorakopoulos, L., Karras, C., Theodoropoulou, A., Kalliampakou, I., & Kalogeratos, G. (2025). LLMs for cybersecurity in the Big Data era: A comprehensive review of applications, challenges, and future directions. Information, 16(11), Article 957. https://doi.org/10.3390/info16110957
24. McIntosh, T. R., Susnjak, T., Liu, T., Watters, P., Xu, D., Liu, D., Nowrozy, R., & Halgamuge, M. N. (2024). From COBIT to ISO 42001: Evaluating cybersecurity frameworks for opportunities, risks, and regulatory compliance in commercializing large language models. Computers & Security, 144, Article 103964. https://doi.org/10.1016/j.cose.2024.103964
25. Mirtaheri, S. L., Movahed, N., Shahbazian, R., Pascucci, V., & Pugliese, A. (2026). Cybersecurity in the age of generative AI: A systematic taxonomy of AI-powered vulnerability assessment and risk management. Future Generation Computer Systems, 175, Article 108107. https://doi.org/10.1016/j.future.2025.108107
26. MITRE. (2024). ATLAS: Adversarial threat landscape for artificial-intelligence systems. The MITRE Corporation. https://atlas.mitre.org
27. Moia, V. H. G., Sanz, I. J., Rebello, G. A. F., de Meneses, R. D., Hitaj, B., & Lindqvist, U. (2026). LLM in the middle: A systematic review of threats and mitigations to real-world LLM-based systems. Computer Science Review, 61, Article 100916. https://doi.org/10.1016/j.cosrev.2026.100916
28. National Cyber Security Centre [NCSC], Cybersecurity and Infrastructure Security Agency, et al. (2023). Guidelines for secure AI system development. NCSC.
29. National Institute of Standards and Technology [NIST]. (2024a). Artificial intelligence risk management framework: Generative artificial intelligence profile (NIST AI 600-1). U.S. Department of Commerce. https://doi.org/10.6028/NIST.AI.600-1
30. National Institute of Standards and Technology [NIST]. (2024b). Secure software development practices for generative AI and dual-use foundation models: An SSDF community profile (NIST SP 800-218A). U.S. Department of Commerce. https://doi.org/10.6028/NIST.SP.800-218A
31. OWASP Foundation. (2023). OWASP Top 10 for large language model applications (Version 1.1). Open Worldwide Application Security Project.
32. OWASP GenAI Security Project. (2025). OWASP Top 10 for LLM and generative AI applications: 2025. Open Worldwide Application Security Project.
33. OWASP GenAI Security Project. (2026). OWASP GenAI LLM Top 10 2026. Open Worldwide Application Security Project. https://genai.owasp.org/resource/owasp-genai-llm-top-10-2026/
34. Page, M. J., McKenzie, J. E., Bossuyt, P. M., et al. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71
35. Pearce, H., Ahmad, B., Tan, B., Dolan-Gavitt, B., & Karri, R. (2022). Asleep at the keyboard? Assessing the security of GitHub Copilot's code contributions. In 2022 IEEE Symposium on Security and Privacy (SP) (pp. 754–768). IEEE. https://doi.org/10.1109/SP46214.2022.9833571
36. Sai, S., Yashvardhan, U., Chamola, V., & Sikdar, B. (2024). Generative AI for cyber security: Analyzing the potential of ChatGPT, DALL-E, and other models for enhancing the security space. IEEE Access, 12, 53497–53516. https://doi.org/10.1109/ACCESS.2024.3385107
37. Schmitt, M., & Flechais, I. (2024). Digital deception: Generative artificial intelligence in social engineering and phishing. Artificial Intelligence Review, 57, Article 324. https://doi.org/10.1007/s10462-024-10973-2
38. Sheng, Z., Chen, Z., Gu, S., Huang, H., Gu, G., & Huang, J. (2025). LLMs in software security: A survey of vulnerability detection techniques and insights. ACM Computing Surveys, 58(5), Article 134. https://doi.org/10.1145/3769082
39. Shenoy, N., & Mbaziira, A. V. (2024). An extended review: LLM prompt engineering in cyber defense. In 2024 International Conference on Electrical, Computer and Energy Technologies (ICECET). IEEE. https://doi.org/10.1109/ICECET61485.2024.10698605
40. Uddin, M., Irshad, M. S., Kandhro, I. A., Alanazi, F., Ahmed, F., Maaz, M., Hussain, S., et al. (2025). Generative AI revolution in cybersecurity: A comprehensive review of threat intelligence and operations. Artificial Intelligence Review, 58, Article 236. https://doi.org/10.1007/s10462-025-11219-5
41. Vadisetty, R., & Polamarasetti, A. (2024). Generative AI-driven distributed cybersecurity frameworks for AI-integrated global Big Data systems. In 2024 International Conference on Emerging Technologies and Innovation for Sustainability (EmergIN) (pp. 595–600). IEEE. https://doi.org/10.1109/EmergIN63207.2024.10961616
42. Wang, X., Tian, Y., Huang, K., & Liang, B. (2025). Practically implementing an LLM-supported collaborative vulnerability remediation process: A team-based approach. Computers & Security, 148, Article 104113. https://doi.org/10.1016/j.cose.2024.104113
43. Xu, H., Wang, S., Li, N., Wang, K., Zhao, Y., Chen, K., Yu, T., Liu, Y., & Wang, H. (2025). Large language models for cyber security: A systematic literature review. ACM Transactions on Software Engineering and Methodology. https://doi.org/10.1145/3769676
44. Xu, Y., Zhuang, Y., Liu, X., Zhang, T., Xiao, B., Xu, X., Jiang, D., Wang, J., et al. (2026). LLM agents security duality: A comprehensive survey of self-security and empowered cybersecurity. Artificial Intelligence Review, 59, Article 174. https://doi.org/10.1007/s10462-026-11563-0
45. Yigit, Y., Buchanan, W. J., Tehrani, M. G., & Maglaras, L. (2025). Review of generative AI methods in cybersecurity. Internet of Things and Cyber-Physical Systems, 5, 241–261. https://doi.org/10.1016/j.iotcps.2026.04.001
46. Zhang, A. K., Perry, N., Dulepet, R., Ji, J., Menders, C., Lin, J. W., et al. (2025). Cybench: A framework for evaluating cybersecurity capabilities and risks of language models. In Proceedings of the International Conference on Learning Representations (ICLR 2025).
47. Zhang, J., Bu, H., Wen, H., Liu, Y., Fei, H., Xi, R., Li, L., Yang, Y., Zhu, H., & Meng, D. (2025). When LLMs meet cybersecurity: A systematic literature review. Cybersecurity, 8, Article 55. https://doi.org/10.1186/s42400-025-00361-w
48. Zhang, R., Li, H.-W., Qian, X.-Y., Jiang, W.-B., & Chen, H.-X. (2025). On large language models safety, security, and privacy: A survey. Journal of Electronic Science and Technology, 23(1), Article 100301. https://doi.org/10.1016/j.jnlest.2025.100301
49. Zhou, W., Zhu, X., Han, Q.-L., Li, L., Chen, X., Wen, S., & Xiang, Y. (2025). The security of using large language models: A survey with emphasis on ChatGPT. IEEE/CAA Journal of Automatica Sinica, 12(1), 1–26. https://doi.org/10.1109/JAS.2024.124983
